Carbon monoxide (CO) leads poisoning deaths worldwide and masquerades as a variety of medical maladies resulting in frequent misdiagnoses.(1) Whether on scene of a general illness call, providing rehab to firefighters at a major fire or responding to a CO detector alarm, EMS providers need more than a list of signs and symptoms to reliably evaluate, treat and safely transport patients affected by carbon monoxide.
This article will review the suspected incidence of CO poisoning and short- and long-term dangers of this invisible poison. It presents a protocol specifically designed for EMS responders using currently available monitoring devices to screen for CO in patients.
The true incidence of CO poisoning isn't known.(1) Recent calculations estimate 50,000 emergency department (ED) visits annually in the U.S. result from CO poisoning, although it’s widely believed that frequent misdiagnosis results in gross under reporting of the actual incidence.(2,3)
CO poisoning is a major public health problem. It accounts for more than half of fatal poisonings in virtually every country worldwide when reported poisoning cases are tabulated.(1) This often surprises health-care providers, who tend to underestimate the prevalence of CO poisoning in their communities.
Where It Comes From
Sources of CO include endogenous exposure (meaning manufactured within the body) and environmental exposures. At the end of the life of red blood cells, their destruction produces CO. As a consequence, all humans continuously exhibit low levels of CO in the blood, called carboxyhemoglobin (COHb). Individuals with hemolytic anemias, sepsis and critical illness have shorter-than-average red blood cell life spans, resulting in higher-than-normal COHb levels.(1)
Environmental carbon monoxide sources encompass any process involving incomplete combustion of a carbon-containing product. These include exhaust from a vehicle, heating appliances, fireplaces, cigarette smoke, barbecue grills, smoke from a fire and a plethora of other combustion-generating devices and processes.(1)
Rarely seen but important to prehospital providers, inhaled or absorbed methylene chloride produces significant amounts of CO that can result in poisoning. Methylene chloride is found in paint stripper and can be a component of chemicals used for “huffing” (deliberate inhalation designed to produce an altered mental state). When broken down in the liver, methylene chloride produces significant amounts of carbon monoxide that can result in CO poisoning.
It has long been mistakenly believed that classic symptoms of CO exposure, such as headache, nausea, weakness and visual disturbances, were reliable indicators of CO poisoning. In fact, signs and symptoms of carbon monoxide poisoning correlate so poorly with COHb levels that no study conducted to date has been able to reliably detect CO poisoning by symptoms or physical assessment findings alone.(1)
The uncanny similarities of CO poisoning to viral illness, gastroenteritis, migraine, acute coronary syndrome or angina, drug abuse and a variety of other medical conditions, coupled with lack of correlation between symptoms and COHb levels, leads to frequent misdiagnosis and has nicknamed CO poisoning as “the great imitator.”(4) In fact, the high incidence of CO poisoning and the difficulty experienced in detecting poisoned patients when a CO monitoring device isn’t available, has prompted many intensive-care unit order sets to include obtaining a COHb level on admission to avoid missing this potential diagnosis.
CO is colorless, odorless, invisible and hence, virtually undetectable. It has a vapor density (0.97) very close to ambient air (1.0). This means it will not linger on the ground or rise upward at usual temperatures, but instead spread throughout a room or building when released.(3)
The lungs readily absorb inhaled CO into the bloodstream where it binds to hemoglobin with substantially greater affinity than oxygen, forming COHb.(1) Hypoxia results both from decreased oxygen-carrying capacity and an increased affinity of remaining oxygen molecules to the hemoglobin. In the presence of CO, remaining oxygen molecules bind more tightly to hemoglobin, refusing to unload into tissues. This phenomenon, which results in higher-than-normal oxygen levels in venous blood, is responsible for the cherry red skin color sometimes seen in CO-poisoned patients.
Hemoglobin delivers CO throughout the body, resulting in a variety of symptoms depending on the concentrations achieved. When CO binds with skeletal muscle, an individual can experience significantly impaired strength that can approach paralysis at high concentrations.
Even low levels of CO produce elevated concentrations of the free radical nitric oxide (NO), a highly reactive chemical compound known to cause significant cellular damage in the body.(5) NO causes vasodilation and leads to profound hypotension in patients with major CO poisoning.
Neurologic & Cardiac Concerns
In addition to the hypoxia and tissue-damaging effects already mentioned, CO also causes inflammation through multiple pathways independent of those associated with hypoxia, leading to direct neurologic and cardiac injury, both with long-term consequences.(4) Therefore, disabling cardiac injury often occurs even with low-level CO exposures, a cause of concern for firefighters who frequently work in smoky conditions and persons who live or work with cigarette smokers.
In a Swedish study monitoring the incidence of cardiovascular disease in 8,333 men over a 19-year period, those who never smoked but had COHb levels consistent with exposure to second-hand smoke were 3.7 times more likely to have a cardiac event (stroke or myocardial infarction) and 2.2 times more likely to die than those who had not been exposed to CO.(6) A particularly disconcerting takeaway from these data is that the COHb levels of men in the higher risk group were very similar to those normally observed in firefighters who, despite proper use of personal protective equipment (PPE), have a certain degree of unavoidable occupational CO exposure.(5)
Numerous studies have documented increased incidences of cardiovascular events and higher mortality in long-term follow up of patients who have been CO poisoned.(7) Therefore, CO-poisoned patients should always receive a thorough cardiac workup.
Troubling neurologic impairment also occurs with CO exposure, leaving up to one-third of CO-poisoned patients with profound neurological disabilities that may include substantial drop in intelligence, memory loss, difficulty concentrating, seizures, Parkinson-like tremors, psychosis, dementia, and cognitive or personality changes.(8)
Catching the Killer
Given the very real and significant incidence of carbon monoxide poisoning in every part of the world, our total inability to detect this very sneaky, colorless, odorless and invisible gas, and the complete lack of relationship between signs or symptoms and COHb levels, the challenge has been how to improve our ability to detect and treat CO-poisoned patients.
With signs and symptoms that range from none to mild headache, tiredness, nausea, vomiting, tachycardia, confusion, seizures and unconsciousness, it would seem that virtually any patient could be poisoned with carbon monoxide. Therefore, to avoid missing this important diagnosis, I believe we need to screen every patient we encounter for carbon monoxide poisoning.
Exhaled breath assessment of CO, used extensively in the UK to monitor compliance with smoking cessation, has been used to assess CO exposure in U.S. firefighters since 1976.(9,10) There is close correlation between exhaled breath CO and COHb, but the technology has never been widely adopted.(11)
Pulse CO-oximetry technology that can non-invasively measure COHb has been available in the U.S. since 2006. Many EMS services and fire departments now carry portable CO-oximeter units. More recently, CO-oximetry has become available in several monitor-defibrillator models.
Conventional pulse oximeters transmit two wavelengths of light (red and infrared) through body tissue, measuring the absorption of each wavelength and calculating the oxygen saturation of hemoglobin (SpO2%). However, because oxygen and CO have similar absorption characteristics, conventional pulse oximetry sees them as identical, reporting an SpO2 falsely elevated by CO.
On the other hand, pulse CO-oximetry uses at least eight wavelengths of both visible and invisible spectrum light to accurately measure COHb, displayed as SpCO%. Its accuracy is reported at plus or minus 3% SpCO at levels up to 40%.(12) Shielding from extraneous light sources and meticulous attention to proper probe placement improves accuracy. Read more about SpCO in "The New Vital Sign Parameter" on p. 24.
In a study to determine what would happen if every patient encountered in the field was screened with pulse CO-oximetry, researchers at Brown University Hospital in Rhode Island found previously unsuspected elevated carbon monoxide levels in four of every 10,000 patients during winter (heating) months and in one of every 10,000 patients during summer months.(13) Although other researchers have questioned these data, the fact remains that CO poisoning is a significant public health problem, and one that frequently eludes detection.(14,15)
Good patient-care decisions flow from good understanding of technology and the knowledge to interpret data obtained from a medical device. Pulse CO-oximetry should never replace clinical judgment; any symptomatic patient should receive further medical evaluation regardless of measured SpCO%. Based on published recommendations and the author’s decade of experience with non-invasive CO assessment technologies, Figure 1 (see p. 5) is a recommended guideline for routine assessment and reassessment of the patient's carboxyhemoglobin levels.(1,12,13,16)
Fire Scene Rehab
An additional and critically important niche for pulse CO-oximetry is firefighter rehabilitation. The National Fire Protection Association (NFPA) 1584 Standard on the Rehabilitation Process for Members During Emergency Operations and Training Exercises suggests that EMS personnel assess any firefighter for CO poisoning who has been exposed to carbon monoxide or presents with symptoms at an incident where CO is present.(17)
When engaged in firefighting, firefighters use self-contained breathing apparatus (SCBA) to protect themselves from respiratory exposure to CO and other toxins known to be present in fire smoke. Yet it's readily apparent that firefighters fail to use SCBA during overhaul and when operating in smoky environments outside or nearby the fire location. These circumstances unnecessarily expose firefighters to CO.(5)
The usual terminal event in CO poisoning is cardiac arrest, most often heralded by ventricular fibrillation.(6,7) You have a clear opportunity to protect firefighters by screening for CO. Screen potentially exposed firefighters for CO, immediately treat any abnormal levels and never release emergency personnel from rehab with an SpCO level greater than 5%.(1)
Given NFPA 1584 and the ready availability of a non-invasive screening tool for detecting CO poisoning, it seems reasonable that EMS providers operating in the rehab area should have this capability available. A sample protocol developed by the author for firefighter rehab is presented in Figure 2 (see p. 7).
Lastly, public health initiatives to reduce the incidence of CO poisoning in the community have enhanced building codes, leading to increasing use of CO detectors in homes and businesses. Responses to CO alarm activations constitute a significant number of fire department runs each year.
Considering the wide variety of conditions that can activate a typical household CO detector, ranging from 30 parts per million (PPM) for 30 days to 400 PPM for 15 minutes, and including other gases such as CO2, methane, and isopropyl alcohol, it's no wonder why firefighters are often unable to determine what tripped the CO alarm.(1)
The goal when responding to a carbon monoxide alarm is to determine whether the environment is poisonous. However, atmospheric monitoring using appropriate gas meters often won't detect low levels of CO, especially when homeowners have exited the building prior to fire department arrival, ventilating CO in the process.
CO-oximetry can play an important role in detecting CO-exposed individuals regardless of atmospheric reading levels obtained on scene, since COHb has a four-hour half life in individuals breathing room air.(16)
Because prolonged exposure to very low levels of CO can be just as lethal as short-term exposure to high CO concentrations, firefighters who screen building occupants with CO-oximetry in addition to monitoring the home or other structure with a four-gas meter, can now more definitively rule out CO poisoning and feel much more confident allowing a family to return to their home. In fact, many fire departments have incorporated protocols, such as the one in Figure 3 (see p. 7), to more confidently respond to CO alarms.
CO is an invisible and insidious poison that often eludes detection by appearing like general illness. We often miss dangerous CO levels because we fail to look for them, chalking up symptoms (when present) to general weakness or flu-like illness. Pulse CO-oximetry now offers us the ability to non-invasively detect CO poisoning during regular and focused patient exams.
To be part of the solution, in my opinion, we need to screen every patient we see, potentially exposed firefighters managed during rehab operations and every building occupant at the scene of a CO detector alarm activation. Using the sample protocols outlined in this article, EMS providers and firefighters have a new opportunity to detect this elusive poison early in prehospital patient encounters and during the performance of rehab operations.
Disclosure: The author has reported receiving honoraria and/or research support, either directly or indirectly, from Masimo.
1. Raub JA, Mathieu-Nolf M, Hampson NB, et al. Carbon monoxide poisoning: A public health perspective. Toxicology. 2000;145:1–14.
2. Hampson NB, Weaver LK. Carbon monoxide poisoning: A new incidence for an old disease. Undersea Hyperb Med. 2007;34:163–168.
3. Kele A, Demircan A, Kurto lu G. Carbon Monoxide poisoning: How many patients do we miss? Eur J Emerg Med. 2008;15:154–157.
4. Weaver LK. Clinical practice. Carbon monoxide poisoning. N Engl J Med. 2009;360:1217–1225.
5. Bledsoe BE. The heart dangers of CO. Understanding cardiovascular risks to responders from CO exposure. JEMS. 2007;32:54–59.
6. Hedblad B, Engström G, Janzon E, et al. COHb% as a marker of cardiovascular risk in never smokers: Results of a population-based cohort study. Scand J Pub Health. 2006;34:609–615.
7. Hampson NB, Rudd RA, Hauff NM. Increased long-term mortality among survivors of acute carbon monoxide poisoning. Crit Care Med. 2009;37:1941–1947.
8. Amitai Y, Zlotogorski Z, Golan-Katzav V, et al. Neuropsychological impairment from acute low-level exposure to carbon monoxide. 1998;55:845-8.
9. Deveci SE, Deveci F, Acik Y, et al. The measurement of exhaled carbon monoxide in healthy smokers and non-smokers. Respir Med. 2004;98:551–556.
10. Stewart RD, Stewart, RS, Stamm W, et al. Rapid estimation of carboxyhemoglobin level in firefighters. JAMA. 1976;235:390–392.
11. Cone DC, MacMillan DS, Van Gelder C, et al. Noninvasive fireground assessment of carboxyhemoglobin levels in firefighters. Prehosp Emerg Care. 2005;9:8–13.
12. Hampson NB. Noninvasive Measure-ment of Blood Carboxyhemoglobin with Pulse CO-Oximetry. In Penney DG: Carbon Monoxide Poisoning. CRC Press: Boca Raton, La, 2008, pp. 739–744.
13. Suner S, Partridge R, Sucov A, et al. Non-invasive pulse CO-oximetry screening in the emergency department identifies occult carbon monoxide toxicity. J Emerg Med. 2008;34:441–450.
14. O’Malley GF. Letter to the Editor: Non-invasive carbon monoxide measurement is not accurate. Ann Emerg Med . 2006;48:477–478.
15. Centers for Disease Control and Prevention. Carbon Monoxide Poisoning. April 2009. www.cdc.gov/co/pib.htm.
16. Crawford DM, Hampson NB. Fire and ice: Diagnosis of carbon monoxide poisoning in a remote environment. Emerg Med J. 2008;25:235–236.
17. NFPA 1584. Standard in the Rehabilitation Process for Members during Emergency Operations and Training Exercises. NFPA: Quincy, Mass., 2008.
This article originally appeared in the October 2010 JEMS supplement “The Silent Killer” as “Sneak Attack: What makes carbon monoxide so insidious?”